review of the major findings about duane retraction ...duane retraction syndrome (drs) is a...

Vision Research 50 (2010) 2334–2347

Contents lists available at ScienceDirect

Vision Research

journal homepage: www.elsevier .com/locate /v isres

Review of the major findings about Duane retraction syndrome (DRS) leadingto an updated form of classification

Demet Yüksel a,b, Jean-Jacques Orban de Xivry c, Philippe Lefèvre b,d,⇑a Ophthalmology Department (St. Luc Hospital and UCLouvain), Brussels, Belgiumb Institute of Neuroscience (IoNS), Université catholique de Louvain, Belgiumc Biomedical Engineering Department, Johns Hopkins University, Baltimore, MD, USAd CESAME and ICTEAM, Université catholique de Louvain, Belgium

a r t i c l e i n f o a b s t r a c t

Article history:Received 18 June 2010

Keywords:DuaneEye movementsStrabismusSaccades

0042-6989/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.visres.2010.08.019

⇑ Corresponding author at: Institute of Neurosciencue de Louvain, Belgium.

E-mail address: [email protected] (P

In view of all the reported evidence by electromyography in the 1970s, by histology in the 1980s, and bycerebral imagery since the 2000s, Duane retraction syndrome (DRS) has been described as the conse-quence of a congenital anomaly of the 6th cranial nerve nuclei with aberrant innervations by supply fromthe 3rd cranial nerve. Both genetic and environmental factors are likely to play a role when the cranialnerves and ocular muscles are developing between the 4th and the 8th week of gestation. New data fromeye movement recordings contributed to better understanding the binocular control of saccades. Model-ing of saccades in DRS seems promising for the quantification of the innervational deficit and themechanical properties of the eye plant. The usual clinical classification of DRS needs to be updated inorder to match more accurately the underlying dysinnervation of the extra ocular muscles and to illus-trate the continuum that exists between the various forms. This review aims to summarize the majorfindings about DRS and to guide the clinician in the surgical management of this particular form ofstrabismus.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction strabismus, this technique provides new insight into the neural

The Duane retraction syndrome (DRS) has driven a lot of re-search in the last decade. Tremendous information about the DRSpathogenesis, its clinical presentation, its treatment and other var-ious aspects of this condition have been reported in the scientificliterature. An analysis of the new available information and ofthe corresponding major findings could lead to a better under-standing of the pathogenesis of this particular form of strabismus.Advances in embryogenesis, in genetics and in electromyographiccharacteristics, plus the visualization of the oculomotor nervesby cerebral imaging lead to accept DRS as being a unilateral orbilateral congenital anomaly of the 6th cranial nerve nuclei withaberrant innervations by supply from the 3rd cranial nerve. Addi-tional information about the underlying innervational deficit hasbeen recently obtained by eye movement recordings. Careful clin-ical evaluation of the eye motility coupled with all collected dataleads to update the previous classification of the different subtypesof DRS in order to guide the surgical management.

Eye movement recording represents a window into the brain asboth neural and mechanical factors determine human eye move-ments (Leigh & Zee, 2006). Therefore, in this particular form of

ll rights reserved.

e (IoNS), Université catholiq-

. Lefèvre).

control of eye movements. The specific anatomical and physiolog-ical substrate of DRS makes it an ideal candidate for studying thebinocular control of saccades, brain adaptation and provides achallenge for testing the existing control system models. Indeed,in contrast with comitant strabismus, binocular vision is preservedin DRS. Indeed, the compensation by abnormal head posture allowsbinocularity in one field of gaze despite the severe eye motility def-icit in the other field of gaze. The degree of sensorial binocular sta-tus plays an important role in the conjugacy of saccades (Kapoula,Bucci, Eggert, & Garraud, 1997). Therefore, unilateral forms of DRSgive the opportunity to study the coupling and/or uncoupling ofthe two eyes (one sound eye and one affected eye) during eyemovements (Yüksel, Optican, & Lefevre, 2005). Binocular record-ings of horizontal saccades to and away from primary positionbrought precious information about the innervational commandsent to the two eyes. The relationship between saccade amplitudeof the affected eye and the sound eye was linear for all directions ofeye movements implying a close coupling between the two eyes. Inaddition, this condition provides another window on the adaptiveprocesses that are required to control the abnormal eye plant. Thecomparative study of saccades in monocular versus binocularviewing conditions enabled to show that monocular adaptationwas possible only for the step of innervation (i.e. controlling the fi-nal eye position) but not for the pulse of innervation (i.e. control-ling the saccadic gain), even though the peculiarity of DRS type I

http://dx.doi.org/10.1016/j.visres.2010.08.019

mailto:[email protected]

http://dx.doi.org/10.1016/j.visres.2010.08.019

http://www.sciencedirect.com/science/journal/00426989

http://www.elsevier.com/locate/visres

D. Yüksel et al. / Vision Research 50 (2010) 2334–2347 2335

offered clear advantage for separate pathways of control for thetwo eyes. This set of data allowed us to propose a model of com-mon pathway for the pulse of innervation for the two eyes withseparate pathways of command for the step of innervation (Yüksel,Orban de Xivry, & Lefevre, 2008). In parallel to yielding insightsinto brain function, eye movements also provide some informationabout the condition itself. For instance, eye movement recording inunilateral DRS during periods of monocular versus binocular visionallows testing for the presence of some residual innervation in theaffected 6th nerve and for the type of transfer of innervation fromone eye to the other according to the viewing condition. In addi-tion, the integrity of the abducens interneurons could be inferredfrom the normal behavior of the sound eye toward the affected-side gaze.

Most of DRS patients compensate well for the disorder and donot require further management. Standard management of DRSmay in some cases involve eye muscle surgery. Surgery does noteliminate the fundamental abnormality of innervation and no sur-gical technique has been completely successful in restoring full nor-mal eye movements in DRS patients. The success rate in eliminatingan abnormal head position is above 80%. Results are stable for atleast 8.75 years after surgery (Barbe, Scott, & Kutschke, 2004). Sur-gery does not normalize horizontal motility. Only transpositionprocedures may in selective cases improve abduction (away fromthe nose = outward rotation of the globe), with some sacrifice onadduction (toward the nose = inward rotation of the globe). Lateralrectus fixation into the lateral orbital wall with augmented transpo-sition of vertical muscles with vessel sparing technique is promis-ing for treating severe DRS associated with abnormal up ordownshoots on adduction (Rosenbaum, 2004). Undercorrectionand overcorrection may occur if the particular mechanical andinnervational aspects of DRS are not taken into account.

In order to optimize DRS management, diagnosis has to bemade based on precise evaluation of the clinical signs. The use ofadditional clinical tests as forced duction, magnetic resonanceimaging, saccade recording and modeling of the saccadic behaviorof DRS may contribute not only to better understand the pathologybut also guide the clinician in the daily practice management andin the surgical planning. In view of various aspects of the patho-genesis, a new form of classification is proposed.

2. General characteristics

Duane retraction syndrome (DRS) is a well-recognized clinicalentity since more than a century. Stilling (1887), Türk (1896) andAlexander Duane (1905) are the early describers of this particularform of strabismus. In European literature the retraction syndromeis appropriately referred to as the Stilling–Turk–Duane syndrome.

DRS is rare in the general population with an incidence of about0.1%. Prior reviews of DRS, comprising mostly unilateral cases, re-port a 1–4% proportion of all strabismus cases. The syndrome isusually unilateral and sporadic; however, numerous cases of famil-ial transmission, mostly bilateral with an autosomal dominantinheritance pattern, have been reported (Sevel & Kassar, 1974). Afamily of 118 members in three generations, including 25 livingmembers affected with DRS in an autosomal dominant patternhas been studied (Chung, Stout, & Borchert, 2000). Studies of DRShave been published in almost all areas of the world, and no partic-ular race or ethnic group presented a predisposition for the syn-drome. The majority of studies pointed out up to 60%predominance of the syndrome among females. This observationled to the hypothesis that the gene was partly sex-limited. The leftside predominance has been cited in all the studies of DRS over thelast century. When all major studies were gathered, over a totalnumber of 835 patients, 59% were left eye affected, 23% were right

eye affected and 18% were bilateral. If bilateral cases were elimi-nated, over a total number of 680 patients, 72% presented a lefteye involvement. There has been no explanation found for this pre-ponderance of left eye involvement (DeRespinis, Caputo, Wagner,& Guo, 1993). Over a total number of 471 patients grouping six ma-jor studies, it appeared that hypermetropia greater than +1.50diopter (D) was more frequent in DRS (71%). Myopia and emmetro-pia appeared in relatively equal amounts (15% and 14%, respec-tively). Anisometropia greater than +1.00 diopter in sphere,cylinder or both was found in approximately 23% of the 471 pa-tients gathered from the major six studies (range of 14–40%). Therange of amblyopia in Duane’s syndrome was from 3% to 25%, withthe weighted average being 14% among studies. Amblyopia wasmainly due to strabismus and not to anisometropia (DeRespiniset al., 1993). Seventy cases of Duane’s syndrome are summarizedwith particular attention to the prevalence of anisometropia andamblyopia (Tredici & von Noorden, 1985). They found a 17% prev-alence of anisometropia and a 3% prevalence of amblyopia amongthese patients. Amblyopia and anisometropia do not seem morecommon in Duane’s syndrome than in the general population. Fix-ation preference usually corresponds to the dominant eye basedessentially on strabismic dominance, and/or the visual acuity.However, a few cases of fixation preference for the affected eyehave been reported (Khan & Oystreck, 2006). DRS is associatedwith various ocular and nonocular malformations (15–50%)according to the time of development of the ocular and nonocularstructures involved (DeRespinis et al., 1993).

3. Central nervous system anomalies

Because DRS is a benign disorder, autopsy subjects are rarelyavailable. Matteuci was the first to report hypoplastic abducensnucleus with absent 6th nerve on the affected side in one DRStype I patient (Matteuci, 1946). The lateral rectus muscle was de-scribed fibrotic and the medial rectus muscle hypertrophic. Theperipheral innervations of the lateral rectus muscle were not dis-cussed, nor were the terminal branches of the oculomotor nervefollowed. Two other autopsy reports have brought neuroanatomi-cal evidence of high importance by complete intracranial and orbi-tal pathologic examination of two cases of DRS in which the clinicalfindings were well documented (Hotchkiss, Miller, Clark, & Green,1980; Miller, Kiel, Green, & Clark, 1982). The first case was bilateralDRS type III and the second was described as DRS type I. Clinicalappearance of the motility of the studied subjects is sure enoughDRS type III for the first report, but is more likely to be a DRS typeII for the second report. Indeed, the limitation in adduction is moresevere than the limitation in abduction. Postmortem examinationof the brainstem and the posterior parts of the orbits revealedbilateral hypoplasia of the abducens nuclei and the abducensnerves. The abducens nucleus contained no motor neuron cellbodies at levels the abducens nuclei normally occupy, but did con-tain several small cell bodies compatible with internuclear neu-rons. No intra-axial fibers referable to the 6th cranial nerve couldbe identified within the brainstem. Both oculomotor nuclei andnerves were normal at the level of the ciliary ganglion. The infe-rior division of the oculomotor nerve divided into severalbranches penetrating the inferior medial aspect of the lateralrectus muscle. Sections through the lateral rectus muscle showhealthy, well-formed muscle bundles in areas innervated by fibersfrom the third nerve. The remaining muscle mass was poorlyinnervated and showed fibrosis. The medial rectus was normal insize and structure. These data brought relevant information to ex-plain major electromyographic findings (Scott & Wong, 1972).

Regarding these elements, it appears that development of thelateral rectus muscle may proceed even if its usual nerve supply

2336 D. Yüksel et al. / Vision Research 50 (2010) 2334–2347

is defective (hypoplasia of the 6th nerve nucleus) because of thepresence of a substitute nerve branch coming from the oculomotornerve. This extra branch may derive from the superior or inferiordivision of the oculomotor nerve. The abducens and the oculomo-tor nerves are closely associated as they pass through the cavern-ous sinus and particularly as they enter the orbit through thetwo heads of the lateral rectus muscle. Anastomoses between thetwo nerves in the cavernous sinus and the orbit provide theoreticalopportunity for intermingling of their axons (Hoyt & Nachtigaller,1965).

4. Embryogenesis

The myofibers of extra ocular muscles (EOMs) of the eye aredeveloped by a condensation of the mesoderm around the eyewhereas muscle connective tissue cells arise from neural crest.Oculomotor nerves grow out from the brain into their respectiveprimordial muscle condensations some time after the latter havebeen formed. When the embryo is 7 mm long, the EOMs formone mass, which is supplied by only the third nerve. When the em-bryo is 8–12 mm long, that is, when the fourth nerve and the 6thnerve arrive, this mass divides into separate muscles. The oculo-motor nerve first appears at the 7–8 mm stage, then the abducensat the 8–9 mm stage and finally the trochlear at the 10–12 mmstage. It is conceivable that, through disturbing influences of un-known origin, branches of the third nerve remain or come into con-tact with the part of the muscle mass which is later to become thelateral rectus. This occurs in the compensation for an aplastic orabsent abducens nerve (Hoyt & Nachtigaller, 1965). Bremer pro-posed that absence of abducens nerve in the human is due to a rel-ative delay in development of the primordial muscles; this allowsthe abducens fibers sprouting from the brainstem to turn caudallyin response to the ‘‘attraction” of the postotic musculature. Subse-quently, the developing lateral rectus muscle is annexed by an-other nerve, the oculomotor nerve. The caudally directedabducens fibers then disappear along with the postotic mesoderm,leaving no trace of their original emergence from the brainstem.Latest studies emphasized the critical role of the protein @2-chimaerin in the developmental function in ocular motor axonpath finding (Miyake et al., 2008).

Given the evidence that DRS results from a maldevelopment ofthe abducens nerve (cranial nerve VI) and that DRS is associatedwith other anomalies in some cases, the syndrome is thought to re-flect a disturbance of normal embryonic development. Either a ge-netic factor or an environmental factor may be involved when thecranial nerves and ocular muscles are developing between the 4thand the 8th week of gestation.

Studies of sporadic forms of DRS showed 10–20 times greaterrisk for having other congenital malformations divided in mainlyfour categories: skeletal, auricular, ocular and neural (Pfaffenbach,Cross, & Kearns, 1972). The skeletal abnormalities involved the pal-ate and vertebral column. The auricular malformations includedthe external ear, the external auditory meatus and the semicircularcanals. Ocular defects concerned the extra ocular muscles and theeyelids including ocular dermoids. Neural defects involved thethird, fourth and 6th cranial nerves. Some other syndromes are de-scribed to be associated with DRS; Okihiro syndrome (forearmmalformation and hearing loss), Wildervanck syndrome (fusionof neck vertebrae and hearing loss), Holt-Oram syndrome (abnor-malities of the upper limbs and heart), morning-glory syndrome(abnormalities of the optic disc or blind spot), and Goldenhar syn-drome (malformation of the jaw, cheek, and ear, usually on 1 sideof the face). Given those clinical data, disturbance between thefourth to tenth weeks of embryogenesis seems most obvious andcould explain the various nonocular and ocular abnormalities in

combination with Duane’s syndrome. A teratogenic event duringthe second month of gestation seems to cause most ocular andextraocular abnormalities observed in combination with DRS. Tha-lidomide has been clearly reported as having a teratogenic effect.

5. Hereditary and genetic factors

Both genetic and environmental factors are likely to play a rolein the development of Duane syndrome. The majority of Duanesyndrome cases are sporadic in origin with only approximately2–5% of patients showing a familial pattern (running in families),and large families are rarely reported. Both dominant forms andrecessive forms of DRS have been documented. In some familieswith dominant DRS, the disease skips a generation (reduced pene-trance) and ranges in severity (variable expressivity). Most familialcases are not associated with other anomalies. Kirkham was thefirst to note a genetic link between the cleft palate, Klippel–Feilanomaly, perceptive deafness and DRS. These anomalies seemedto be manifestations of a pleiotropic gene inherited in an irregu-larly dominant manner (Kirkham, 1969; Kirkham, 1970a, 1970b).Studies of monozygotic twins have added some confusion to thegenetics of DRS. Some reported concordant cases (Hofmann,1985), some reported discordant cases (Kaufman, Folk, & Miller,1989; Rosenbaum & Weiss, 1978) and one study reported monozy-gotic twins with unilateral DRS, each with the opposite eye af-fected, described as ‘‘mirror images” (Mehdorn & Kommerell,1979). There is currently no test that can determine whether a pa-tient has a hereditary form.

Genetic linkage studies of a large family with DRS establishedthe location of a DRS gene on chromosome 2. Although a geneticcause of DRS has long been accepted, these studies were the firstto show a statistically significant linkage. Only one genetic locusfor isolated DRS has been established by linkage analysis: theDURS2 locus on 2q31 (Appukuttan et al., 1999; Evans, Frayling, El-lard, & Gutowski, 2000). Cytogenetic results (a study of chromo-somes) of individuals with Duane syndrome have, in rare cases,shown abnormalities that suggest that there may be more thanone gene responsible for causing DRS. Deletions of chromosomalmaterial on chromosomes 4 and 8, and the presence of an extramarker chromosome thought to be derived from chromosome 22have been documented in DRS individuals.

Observations led to the feeling that various internal stimuliduring embryogenesis determine uni-or bilaterality, sidenessand degree of expression. The involved gene was proposed tobe incompletely penetrant with variable expressivity. Identifica-tion of the genes mutated in inherited DRS can provide insightboth into the cause of the disorder and the molecular pathwaysessential to ocular motoneuron and axon development. Using thisapproach several mutations in the transcription factor SALL4 havebeen identified to cause DRS in association with variably pene-trant radial ray deformities and deafness gene defects that resultin Duane-radial ray syndrome (DRRS), a disease that was found tomap to Chromosome 20 (Engle, Andrews, Law, & Demer, 2007).The finding that the DRS phenotype maps to the DURS2 locushas provided an opportunity to define further the DURS2-linkedDRS phenotype. Comparison of the clinical and MRI findings with-in DURS2-linked DRS families and the sporadic DRS provide guid-ance for future examinations of the role of the DURS2 gene inocular motor development. DURS2-linked DRS has been reportedto be a diffuse congenital cranial dysinnervation disorder not lim-ited to the abducens nucleus and 6th cranial nerve (Demer, Clark,Lim, & Engle, 2007a). Similar to congenital fibrosis of the extraocular muscles (CFEOM), DRS may be classified as strabismus, un-der the subclassification of incomitant strabismus and extraocularmuscle fibrosis syndromes. Although the term muscle fibrosis


suggests that syndromes under this heading are primary disor-ders of muscle, evidence suggests that DRS (and other syndromesunder this heading, including CFEOM) may be primary disordersof nerve innervation. A most recent study reported that @2-chimaerin has a critical developmental function in ocular motoraxon path finding because expression of mutant @2-chimaerin con-structs in chick embryos resulted in failure of oculomotor axons toinnervate their target extra ocular muscles (Miyake et al., 2008).Genetic studies are promising for the future and new classificationsmight be conducted in view of innovative genetic findings.

6. Clinical description of Duane retraction syndrome (DRS)

The detailed clinical description of the oculomotor disorder hasbeen reported with the following manifestations:

(1) complete, or less often partial, absence of outward movement(abduction) of the affected eye (Fig 1A);

(2) partial, or rarely complete, deficiency of inward movement(adduction) of the affected eye (Fig 1B);

(3) retraction of the affected eye into the orbit when it isadducted (Fig 1B);

(4) partial closure of the eyelids (pseudoptosis) of the affected eyewhen it is adducted (Fig 1B);

(5) a sharply oblique movement of the affected eye, either downand in (downshoot) or up and in (upshoot, Fig 1C), when it isadducted;

(6) paresis, or at least marked deficiency of convergence, with theaffected eye remaining fixed in the primary position whilethe other eye is converging; (Fig 1D);

Fig. 1. (A) Gaze to the left (affected-side gaze) of patient JLB with unilateral DRS type I. LeDRS type I. Left eye is affected. There is partial deficiency of inward movement (adductiowhen it is adducted. (C) Gaze to the right (sound-side gaze) of patient LT with unilateralaffected eye when it is adducted, so-called ‘‘upshoot”. (D) Patient PM shows unilateral DRand vergence of the affected eye (left eye) is limited (bottom). (E) Patient JB turns the headwith half of the face less developed on the left side. Diagnosis is unilateral DRS type I o

(7) abnormal head posture is adopted to search for the directionof gaze where there is no misalignment of the two eyes inorder to obtain binocularity. Longstanding torticollis (sincebirth) leads commonly to craniofacial asymmetry (Fig 1E).For clinicians, the sign of facial symmetry is precious to differ-entiate congenital from acquired forms of motility disorder.

In the clinical setting, the principal difficulties in differentialdiagnosis arise as a consequence of the very early age at which pa-tients with this condition first present. The clinician must be verycareful in examining abduction and adduction, as well as in lookingfor any associated palpebral fissure changes or head postures,when attempting to determine whether what often presents as acommon childhood squint is, in fact DRS. Some cases are challeng-ing because the palpebral signs are uncertain. Some additionalclinical signs have to be pointed out for deciding whether anyabduction limitation is the result of DRS and not a consequenceor abducens cranial nerve palsy. The following clinical signs evokeDRS and not abducens nerve palsy:

(1) Despite the severe limitation of abduction, there is no severeeye misalignment in primary position in DRS (Fig 2A) con-trary to 6th nerve palsy which results in large angle of eso-tropia in primary position (Fig 2B).

(2) Deficiency of abduction is less pronounced in elevation anddepression in DRS (Fig 3A) while the limitation of abductionremains the same in elevation and depression in 6th nervepalsy (Fig 3B).

Disorders similar in clinical presentation to DRS have been de-scribed as a result of trauma (Duane, Schatz, & Caputo, 1976), orbi-

ft eye is affected. (B) Gaze to the right (sound-side gaze) of patient JLB with unilateraln) of the affected eye with retraction into the orbit and partial closure of the eyelidsDRS type I. Left eye is affected. There is a sharply oblique upward movement of theS type I. Left eye is affected (top). Vergence of the sound eye (right eye) is completeto the left for searching eye alignment in right gaze. Facial asymmetry is noticeable

n the left side.

Fig. 2. (A) Patient JLB is orthotropic in primary position with the head straight (top), despite severe limitation of abduction of the left eye (bottom). Diagnosis is DRS type I onthe left eye. (B) Patient FL is esotropic in primary position (top) secondary to absence of abduction of the left eye (bottom). Diagnosis is complete 6th nerve palsy secondary toneurosurgical tumor excision for therapeutical purpose.

Fig. 3. (A) Patient JLB presents DRS type I of the left eye with severe limitation of abduction in horizontal gaze to the left (center). The limitation of abduction improves whenthe eye looks to the left and up (top) and when the eye looks left and down (bottom). (B) Patient FL presents complete 6th nerve palsy of the left eye with complete absence ofabduction in horizontal gaze to the left (center). The limitation of abduction does not improve when the eye looks to the left and up (top) and when the eye looks left anddown (bottom).


tal surgery (Sood, Srinath, & Krishnamurthy, 1975), pterygiumexcision surgery (Khan, 2005), following localized infection (Mur-thy, 2008), neoplasm (Kivlin & Lundergan, 1985) or inflammationof the orbital soft tissue and/or of the extra ocular muscles leadingto consequent mechanical restrictions of eye movement (Kargiet al., 2005; Moorman & Elston, 1995; Osher, Schatz, & Duane,1980). Neurogenic causes that mimic Duane-like syndrome havealso been reported in the literature after surgery on the 5th cranialnerve (Smith & Damast, 1973) or secondary to pontine glioma (Aki-ko, Masato, Tetsuro, & Keichi, 2003). The retraction of the globe wasthe main common picture and occurred either in adduction or inabduction associated to horizontal motility limitation. But in allcases, the underlying cause was different from DRS and thereforethe name ‘‘pseudo-Duane’s retraction syndrome” was preferred.

‘‘Vertical retraction syndrome” indicates the presence of retrac-tion movements in vertical gaze (Khodadoust & von Noorden,1967; Prakash & Menon, 1981; Spielmann, 1988). Because of theiranalogy with Duane’s syndrome, the vertical restriction syndromeshave sometimes been called ‘‘vertical Duane’s syndrome.” This ter-minology should be avoided and substituted by the term ‘‘pseudo-vertical” Duane’s syndrome to avoid confusion.

It is however clear that retraction syndrome is a wide spectrumof cranial dysinnervation that may be spread to cranial nervesother than the abducens. DRS should be utilized only when the lat-eral rectus is concerned. The understanding of the underlyingmechanism is essential to avoid misdiagnosis and confusion.

7. Electromyographic findings

From as early as 1900–1957, numerous investigators reported aspectrum of anatomical findings when operating on extra ocularmuscles (Breinin, 1957; Gobin, 1972). Studies led to the conclusionthat DRS was a local, purely myogenic phenomenon. Posteriorly ormultiply inserted medial rectus muscle and/or fibrotic, inelasticlateral rectus muscle have been widely described. It was believedthat the cause of abduction deficiency was the fibrosis of the lateralrectus muscle and that limitation of adduction was due to theabnormal posterior insertion of the medial rectus muscle or toadhesions between the medial rectus muscle and the orbital wall.Other surgeons found no macroscopic anomalies of both horizontalrectus muscles and assumed a faulty action of the vertical musclesor a deficiency of the check ligaments. All these findings could ex-plain some of the characteristics of the ocular motility but failed toexplain all the components of the clinical entity.

Electromyographic investigations allowed the recording of theelectrical activity in human extra ocular muscles for different eyepositions. Gaze-directed extraocular muscle innervations were re-corded. Breinin (1957) used electromyography and was the first todescribe the absence of potentials of the lateral rectus muscle inabduction, but the presence of action potentials in adduction (Bre-inin,1957). This finding was the evidence of co-contraction of themedial and lateral rectus muscles in adduction and was maderesponsible for the observed globe retraction in adduction. There


were many other electromyographic studies to follow. The actionpotentials of the lateral rectus muscles ranged from no activityduring abduction to essentially equal activity in adduction versusabduction. Medial rectus activity was apparently normal. By1974, Huber gathered all electrophysiological information to con-clude that a paradoxical innervation of the lateral rectus muscleof the affected eye represents the pathogenic principle of all theretraction syndromes (Huber, 1974). The consequent disorder ofthe agonist–antagonist interaction between the medial and lateralrectus muscles has been explained to be responsible for the ob-served secondary anatomical alterations in the muscles. With re-gard to the various electromyographic patterns, Huber’sclassification system groups DRS into three entities; type I (lim-ited abduction with normal to near normal adduction), type II (lim-ited adduction with normal to near normal abduction), type III(limited abduction and adduction). This scheme has been widelyaccepted for clinical and electromyographical classification. Overa total number of 340 patients gathering five major studies, a threeto one preponderance of DRS type I (according to Huber’s classifi-cation) was noted: DRS type I (78%), DRS type II (7%) and DRS typeIII (15%) (DeRespinis et al., 1993). A modification of Huber’s clas-sification was proposed by Ahluwalia, Gupta, Goel and Khurana(1988). It brought many practical clinical aspects by includingthe deviation noted in the primary position of gaze in each ofthe three groups. It is only relevant for DRS type I because DRStypes II and III always have exotropia in primary position (pp).The proposed classification is as follows: DRS type IA (esotropiain pp), DRS type IB (exotropia in pp) and DRS type IC (orthotropiain pp).

The electromyographic classification of DRS has been widelydiscussed and completed. Different types of anomalous lateral rec-tus muscle innervations in DRS have been recorded since Huber.Even in the subgroup of DRS type I, it appeared to be various inner-vations patterns. This information should be brought to light be-cause its implication in the prognosis of the surgicalmanagement is very important. A large review of different patternsof the electrical activity measured in the affected lateral rectus (LR)depending on the gaze has been extensively described by Jampol-sky (1999). Aoki and Mukuno (1989) reported the existence ofthree EMG subtypes in clinical DRS type I. Mizukawa and col-leagues have recently described two additional new types of elec-tromyographic recordings (Mizukawa, Kimura, Fukai, & Tabuchi,2004). These observations bring some new information correlatedwith subgroups of DRS type I and match better the clinical patternin DRS type I with various abnormal head postures and differentdeviation of the eyes in primary gaze.

In view of the tremendous information that the electromyogra-phy has provided, horizontal limitation of abduction and adductionto varying degrees has been well explained by the abnormal inner-vation of the lateral rectus, so-called paradoxical innervation. Med-ial rectus muscle innervation was always normal. The globeretraction and the narrowing of the lid fissure in adduction havebeen accepted to be secondary to the co-contraction of both horizon-tal muscles in adduction. The rearward force to generate globeretraction was measured, and both the co-contraction theory andthe lateral rectus fibrosis theory seemed to be capable of explain-ing the phenomenon (Scott & Wong, 1972). Explanations for up-shoots and downshoots of the affected eye in adduction havebeen strongly disputed between mechanical and anomalous inner-vation theories. The ‘‘bridle effect” in the horizontal muscles sec-ondary to fibrosis and co-contraction would create a certainamount of slippage toward elevation or depression (Miller,1989). This theory was widely supported by the success of surgicalweakening procedures and intramuscular injections of xylocaïneperformed on the horizontal muscles (Magoon, Cruciger, Scott, &Jampolsky, 1982; von Noorden & Murray, 1986). In contrast, the

same procedures performed on vertical and oblique muscles failedto eliminate these vertical abnormal movements. However, ante-rior views near the level of the rectus pulleys showed no evidenceof horizontal rectus EOM sideslip, indicating that the downshootwas not due to a ‘‘bridle effect” during horizontal rectus co-con-traction (Demer et al., 2007a). But, so much improvement in upor downshoot movements has been obtained by surgical proce-dures performed on horizontal muscles, one would imagine thateven if the pulleys do not show sideslip, the restrictive phenomenamay occur more anteriorly and concern the anterior insertions ofthe EOMs. Electromyography (EMG) has certainly clarified the nat-ure of the lateral rectus muscle anomalous innervation but fewmisinterpretations and artifacts have led to some confusion con-cerning supplied innervation from the oculomotor nerve. The ma-jor problem with electromyographic studies consists in theuncertainty about the site of recording in the muscle fibers. Thisconcerns especially the lateral rectus muscle which is composedof various zones of residual muscle fibers due to the innervationeither by the supplying 3rd nerve, either by residual tiny 6th nervefibers and zones of atrophy with fibroblastic tissue secondary toabsence of any innervation (Scott & Wong, 1972). The waveformof the recording depends on the site of implementation of the nee-dle. Moreover, there is no quantitative correlation between theamplitude of the signal and the range of the eye movement. Elec-tromyographic recordings during a saccade show a large motorunit input. Studies measure the signal when the eye is maintainedin an eccentric position. This does not represent the signal of inner-vation as the pulse and step signal sent to the eye muscle. There-fore eye movement recordings are interesting because they givemore information about the motor command sent to the eye mus-cles. Indeed, the study of saccades infers information about the sig-nal of innervations and the mechanical properties of the eyemuscles. However, the information gathered by eye movementrecordings does not give information about one extraocular musclebut concerns the resultant of the activity of all the extra ocularmuscles during a particular gaze, thus it will usually prove veryuseful to combine a variety of recordings in different conditionsin order to make sound conclusions.

8. Data from cerebral and orbital imaging

Magnetic resonance imaging (MRI) appeared to be a preciousnew tool to visualize the brain and the pathway of the cranialnerves over the past 20 years. Indeed, the new techniques offerhigh resolution images in a non-invasive way and the contributionof MRI to the clinical management of strabismus and complex ocu-lomotor disorders has been undeniable. Motion-encoded magneticresonance imaging (MRI) was recently used for the study of humanextraocular muscle (EOM) function; local physiologic contractionand elongation (deformation) were quantified (Piccirelli et al.,2009).

The visualization of the abducens nucleus itself at a neuronal le-vel remains unfeasible. But, the nerve can be explored at the pon-tomedullar level. MRI in cases of DRS type I, demonstrated theabsence of the abducens nerve (Parsa, Grant, Dillon, du Lac, &Hoyt, 1998; Yüksel et al., 2005). This observation is illustrated byan axial-oblique reformatted T2-weighted image of the brainstemat the pontomedullar level of a subject presenting a unilateral formof DRS type I on the right side (Fig 4). However, the variety of thepresence of the abducens nerve in DRS was pointed out by onestudy (Ozkurt, Basak, Oral, & Ozkurt, 2003). These results mustbe interpreted with caution. Not only did they not observe theabducens nerve in one of the 16 eyes in a control group, but alsothere was no classification of the types of DRS patients. More re-cently, a study compared MRI findings of 23 DRS patients with a


control group of 20 individuals. Results confirmed the absence ofthe abducens nerve on the affected side in all patients with DRStype I, mostly in all patients with DRS type III, while DRS type IIwas usually associated with normal 6th nerve (Kim & Hwang,2005b; Yüksel et al., 2005). The most recent MRI study in DRS de-scribed two cases of DRS with absence of the abducens nerve inboth cases. The classification of DRS reported one DRS type I andone DRS type II. The authors suggested that DRS type II could alsobe associated with abducens nerve aplasia. However, the patientclassified DRS type II seems to be more intermediate betweentypes IB and II. Indeed, in the patient described as DRS type II, evenif adduction was more limited than abduction, the deviation in pri-mary position was ortho to small angle exotropia which is moreequivalent to the subgroup type IB. It is therefore essential to cor-relate MRI findings with clinical data. This heterogeneity of the re-sults regarding the visualization of a partial or complete absence ofthe abducens nerve raises the question about the importance of theclassification of the syndrome based on its pathogenesis. In fact,histological studies described above concerned only patients withDRS types I and III. DRS type II is extremely rare and therefore re-quires more cases than already reported for better understandingits innervational characteristics. An interesting report comparedMRI findings in patients with congenital fibrosis of the extra ocularmuscles (CFEOM = congenital ocular motility disorder that mani-fests as restrictive ophthalmoplegia with ptosis) and patients withsynergistic divergence (variant of DRS); the oculomotor nerve washypoplastic bilaterally, and the abducens nerve was absent on theside exhibiting synergistic divergence (Kim & Hwang, 2005a).

As to the visualization of the peripheral abnormal innerva-tions by MRI, coronal T1-weighted magnetic resonance images ofthe orbits and heavily T2-weighted images in the plane of the cra-nial nerves were obtained in subjects with DRS linked to DURS2 lo-cus on chromosome 2 (Demer et al., 2007a) and in subjects withDuane-radial ray syndrome (DRRS) caused by mutations in thetranscription factor SALL4 (Demer, Clark, Lim, & Engle, 2007b).The lateral rectus is shown to be co-innervated by the third cranialnerve in both reports. DRS linked to DURS2 locus is described to beassociated with anomalies of many orbital motor nerves and struc-tural abnormalities of all EOMs except those innervated by theinferior branch of the third nerve contrarily to DRRS for whichthe EOMs and pulleys were structurally normal in most subjects.Although MRI resolution is inadequate to confirm the presence ofmotor endplates in the LR, the inferior division of CN3 was re-ported to run consistently adjacent to the deep portion of the LR

Fig. 4. Axial-oblique reformatted T2-weighted image of the brainstem at thepontomedullar level in one subject presenting unilateral DRS type I on the rightside. Right arrow: left 6th nerve at its emergence from the inferior pons in thesubarachnoid space. Left arrow: clear absence of the 6th nerve at the same level.Adapted from Yüksel et al. (2005).

where CN6 normally enters and arborizes. LR muscles regionalabnormalities were described in subjects with DURS2, in the formof hypoplasia and absent contractility in the deep portion of theEOM belly. Where present, CN6 was seen to enter the superiorzone of the LR, whereas aberrant innervation from CN3 enteredthe inferior zone. This finding was reported as implying that theLR in DURS2 is a two-headed EOM, with each head separatelyinnervated but joined to a common scleral insertion. This func-tional anatomic observation brings explanation to the previouselectromyographic observations in DRS of evidence of two popula-tions of LR motor units: the presumably normally innervated pop-ulation activated during abduction, and an abnormally innervatedpopulation originally destined for the medial rectus (MR) (Scott &Wong, 1972). These two populations presumably represent thetwo zones of the LR visualized by MRI.

The cross-sectional area of extra ocular muscles (EOMs) in dif-ferent directions of gaze (primary position, affected-side gaze,sound-side gaze) is illustrated by axial and coronal T1-weightedimages of the orbits in one subject presenting unilateral DRS typeI on the right side (Fig 5). Images were interpreted in correlationwith the knowledge of the innervational input necessary in conju-gate eye movements for which the abducens nucleus is the mainactor and is the main cause of the disorder in DRS (Yüksel et al.,2005). In this way, MRI brought information about the paradoxicalinnervation of the affected lateral rectus in DRS. The first findingconcerned the absence of atrophy of the lateral rectus muscle bodyin the affected eye in primary position where the eyes were alignedduring constant fixation, despite the absence of innervation by the6th nerve on that side. It is in accordance with other imaging stud-ies of DRS type I. A denervated muscle usually atrophies. The LRmuscle exhibits profound atrophy in severe abducens palsy (Kang& Demer, 2006). Indeed in partial abducens nerve palsy, the extra-ocular muscle size may not demonstrate significant muscle atro-phy (Ozkan & Aribal, 2007). The sparing of the LR in Duanesyndrome from denervation atrophy despite absence of normalabducens innervations suggests existence of alternative LR inner-vation. The importance of this observation is illustrated by a reportof an unusual case of Duane’s syndrome who presented with recur-rent, large-angle esotropia and uncharacteristic atrophy of the lat-eral rectus muscle on magnetic resonance image (MRI) scan(Silverberg & Demer, 2001). The atrophy of the lateral rectus mus-cle was explained by the presence of a skull base meningiomawhich presumably was responsible for compression of anomalousbranches of the oculomotor nerve. Imaging of LR by high-resolu-tion MRI thus appears to be a useful diagnostic tool for discriminat-ing Duane syndrome type I from chronic abducens palsy. It mightbe of particular clinical use in children, and/or in challenging clin-ical patterns. The second finding concerned the visualization of LRand MR co-contraction of the affected eye during sound-side gazemovements. Indeed, the cross-sections of the muscle body of thelateral and the medial recti are equally large without any elonga-tion of the lateral rectus muscle indicating a co-contraction phe-nomenon. This brought an indirect evidence of anomalousinnervation of the affected lateral rectus by the ipsilateral thirdnerve. The absence of LR atrophy is therefore explained by someother innervation arising from the oculomotor nerve which pre-vents the LR from denervation atrophy. Since then, new imagingstudies have reported direct visualization of the co-innervation ofthe affected lateral rectus muscle by branches of the oculomotornerve. Comparison of these findings with the results of orbitalimaging in other fibrosis syndromes (CFEOM) raises the questionof including DRS in what would be called in more general way dys-innervation syndromes.

In conclusion, the high performance of the new imaging tech-niques by MRI certainly confirmed the maldevelopment of theabducens nerve in DRS and showed the compensatory innervation

Fig. 5. Axial (A–C) and coronal (D–F) T1-weighted images through the orbits in one subject presenting unilateral DRS type I on the right side, obtained during constantfixation to affected-side gaze (A and D), in primary position (B and E), and to sound-side gaze (C and F). The contour of the EOMs was manually drawn on coronal viewsthrough the midline portion of the EOMs to visualize the relative movements of the globes and the relative sizes of the EOMs in each eye, in each direction of gaze. Adaptedfrom Yüksel et al. (2005).


by the third nerve at a peripheral level. In this regard, MRI becamean extremely precious tool for guiding the diagnosis and the man-agement of difficult clinical features of oculomotor disorders. How-ever, information about the brainstem circuitry at a neuronal levelstill remains impossible. Therefore, the adjunction of other tech-niques is useful. Eye movement recordings are an additional toolfor understanding the underlying pathogenesis of DRS.

9. Data from eye movement recordings

A technique that could provide quantitative information aboutthe strength of an extraocular muscle would be a useful additionto our clinical evaluation of patients with strabismus. Informationwould be available to assist the diagnosis in difficult cases and aidin management decisions. Direct measurements of muscle forceare difficult to perform quantitatively; they are usually not possi-ble in the paediatric age group and not comfortable in adult pa-tients. In contrast, eye movements can be measured relativelyeasily and accurately with equipment available in many hospitalsand medical centers.

Eye movement recording is a non-invasive technique that hasgiven valuable information about the neural control of movement(Clarke, Ditterich, Druen, Schonfeld, & Steineke, 2002; Collewijn,van der Mark, & Jansen, 1975; Leigh & Zee, 2006; Robinson,1963). Saccades are fast eye movements (up to 500�/s) that enableus to rapidly redirect our line of sight (fovea) toward the object of

interest (Leigh & Zee, 2006). Saccades are characterized by a con-sistent relationship between their peak velocity and their ampli-tude, called the main sequence. Given that these relationships areall fairly stereotyped, one can use quantitative measurements ofsaccades to assess the function of the oculomotor system.

The kinematics of a saccadic eye movement is directly related tothe force produced by extra ocular muscles and is thus an indicatorof the strength of the muscles and their innervation. Therefore, themeasurement of saccadic movements can provide an objective testto evaluate rectus muscle function. Moreover, it gives insight tocentral oculomotor organization. Many previous studies of DRShave provided qualitative descriptions of eye movements and in-volved a mix of all types of patients with DRS (types I–III). Never-theless, in most studies, the properties of saccades were analyzedwithout making any distinction between centrifugal and centripe-tal movements, thus ignoring the manifest asymmetry in DRS.Most studies used low resolution techniques, such as electro-ocu-lography (Gourdeau, Miller, Zee, & Morris, 1981; Metz, Scott, &Scott, 1975; Nemet & Ron, 1978; Prieto-Diaz, 1985). Moore, Feldonand Liu (1988) were the first to perform high-resolution recordingsin two patients with DRS type I. In general, their findings supportedthe hypothesis that DRS results from a central reorganization of theocular motor function in the brain stem. However, they reportedno differences between the dynamic characteristics of adductingsaccades in the affected eye and those of adducting or abductingsaccades in the sound eye, which contradicted all previous reports.This unusual finding is probably due to the pooling of centrifugal


and centripetal movements in their data analysis. In general, allstudies reported marked reduction in abduction saccadic velocityin type I Duane’s syndrome. This is explained by the ocular electro-myographic evidence of lack of muscle activity or recruitment ofthe lateral rectus muscle on attempted abduction. However, theinability to abduct the affected eye might be due in part to restric-tion by stiff medial tissues. This can be tested by force generationtests (Scott, 1971, 1975). Adduction saccadic velocities were mod-erately decreased. Ocular electromyography indicated that the

Fig. 6. Typical traces of centripetal eye movements toward the affected-side gaze, reconditions (BV, SEV, AEV). Positive horizontal eye position corresponds to the affected-sitypical binocular viewing (BV) recording for subject MB and LT. (B) A typical sound eye vierecording for subject MB and LT with behavior of saccadic gain >1. (D) A typical AEV recorof the figure illustrate the target movement in centripetal direction from sound-side gazfixates the target in an eccentric position (left dot). Target is extinguished and appearssaccades for the affected eye and centripetal adducting saccades for the sound eye. Seainclude; BV = binocular viewing (both eyes open), AEV = affected eye viewing (sound eyeviewing eye to track the target while eye movements were measured simultaneouslyrectangle lying before the occluded eye. Adapted from Yüksel et al. (2008).

medial rectus recruits and inhibits normally in Duane’s syndrome.It was suggested that the reduced saccadic velocity on adduction isdue to the co-contraction. This differentiates type I Duane’s syn-drome from lateral rectus palsy. In one DRS type I patient, horizon-tal saccades were measured before and after lateral rectusrecession of the involved eye (Metz, 1983). Abduction saccadeswere unchanged, while adduction saccades improved to normal.The lateral rectus muscle, which had been co-contracting onadduction, had less effect following recession, with resultant

corded with the search coil technique, for two subjects and in different viewingde gaze. Negative horizontal eye position corresponds to the sound-side gaze. (A) Awing (SEV) recording for subject MB and LT. (C) A typical affected eye viewing (AEV)ding for subject MB and LT with a staircase behavior. The diagrams on the right part

e (left on the diagram) toward the affected-side gaze (right on the diagram). Subjectin the center fixation position (right dot). The task induces centripetal abducting

rch coils measure eye position of the two eyes simultaneously. Viewing conditionsoccluded) and SEV = sound eye viewing (affected eye occluded). Subjects used the

in the viewing and the non-viewing eye. The eye patch is represented by a black


improvement in the speed of nasal movements. The importance ofidentifying anomalous co-contraction of the lateral rectus musclebefore planning an operative procedure has been pointed out(Blodi, Vanallen, & Yarbrough, 1964). Indeed, the results of trans-position surgery of the vertical muscles toward the lateral rectusmuscle seem unpredictable and present a high percentage ofovercorrection.

Systematic investigation of horizontal saccadic eye movementsin DRS with high resolution eye movement recording for accuratemeasurements and precise quantification of the metrics and kine-matics of saccades has been useful for improving the understand-ing of the binocular control of saccades (Yüksel et al., 2005). In thisstudy, the two eyes were recorded simultaneously in binocularviewing condition. Saccade properties were quantitatively ana-lyzed separately in four different categories: centrifugal and cen-tripetal saccades for each side of gaze (affected side and soundside). This approach made possible the interpretation of the under-lying innervations and the agonist–antagonist muscle relationshipin order to attempt modeling of saccades in DRS type I. Particularlyin DRS, the initial position of the two eyes is different depending onthe category of eye movement (for instance, when saccades to thesound side are considered, the initial position of the two eyes is dif-ferent for centripetal movements whereas it is the same for centrif-ugal movements). In view of the description of the many varioustypes of DRS, it appeared to be very important to study patientswith very similar clinical pattern because it emphasized the under-lying innervational anomaly. Unilateral DRS type I was interestingbecause it allowed comparing the affected eye behavior with thesound eye. Saccades of the sound eye were orthometric and accu-rate with no drift in centripetal and centrifugal direction. This nor-mal behavior was an evidence of the integrity of the interneuronsin the abducens nucleus. In view of these results, DRS needs to beconsidered as a maldevelopment of the abducens motoneurons andnot of the abducens nucleus. These findings illustrated a mismatchbetween the pulse and step of innervation. The relationship be-tween saccade amplitude of the affected eye and the sound eyewas linear for all categories of eye movements implying a closecoupling between the two eyes. Results brought evidence for con-jugate adaptation of the pulse signal of innervation which seemedto be yoked for the two eyes even though DRS offers clear advan-tage for independent control. In contrast, separate adaptation ofthe step signal of innervation was possible for the two eyes.

Binocular recordings permitted the simultaneous observation ofthe two eyes which present in DRS very different ranges of ocular

Fig. 7. The diagram represents the relationship between the saccadic amplitude in the AEeach eye (left, subject MB; right, subject LT). Means of saccadic amplitudes for each eye itarget step amplitude. For each bin of step amplitude, bi-directional 95% confidence inteAEV conditions. Adapted from Yüksel et al. (2008).

rotations. The comparison between binocular viewing and monoc-ular viewing (sound eye viewing or affected eye viewing) experi-ments was essential for gathering information about theadaptation of the brain. The binocular coordination of saccades isessential for achieving binocular vision after every change in fixa-tion. In addition, binocular vision is necessary to ensure binocularoculomotor coordination between the two eyes via adaptive mech-anisms. Contrary to comitant strabismus, unilateral DRS type I ispeculiar because binocular vision is preserved in one half of the vi-sual field, while there is congenital severe eye misalignment in theother half of the visual field. The coupling of horizontal saccades ofthe two eyes and their dynamic behavior under binocular andmonocular viewing conditions in patients with unilateral DRS typeI was reported (Yüksel et al., 2008). Typical traces of centripetal eyemovements toward the affected-side gaze, recorded with thesearch coil technique, for two subjects and in different viewingconditions (binocular viewing: BV, sound eye viewing: SEV, af-fected eyeviewing: AEV) are illustrated in Fig 6. The comparisonbetween binocular viewing and monocular viewing conditionsbrought more evidence for validation of the hypothesis that the ra-pid part of the saccadic command is common for the two eyes.First, affected eye viewing conditions experiments assessed if therewas some possibility for increasing the innervation (pulse signal ofinnervation) of the affected eye in affected-side gaze searching forthe existence of some residual innervation of the deficient abdu-cens nerve. Moreover, analysis of the results demonstrated thatthere was transfer of the adaptation of the affected eye (increaseof pulse signal) toward the sound eye (occluded). The relationshipbetween the saccadic amplitude in the AEV condition and the sacc-adic amplitude in the BV condition for each subject and for eacheye is illustrated in Fig 7. Finally, depending on the type of adapta-tion (the pulse and/or the step component of saccadic command)and the presence of the transfer to the other eye, we shed lighton saccadic adaptation mechanisms in a theoretical framework.For saccades, the innervation results from a combination of thepulse and the step signal. In conclusion, monocular adaptationwas shown to be possible only for the step of innervation (i.e. con-trolling the final eye position: monocular adaptation of the step,see gain K in Fig 8) but not for the pulse of innervation (i.e. control-ling the saccadic gain: yoked adaptation for the pulse, see gain KAEV

in Fig 8), even though the peculiarity of DRS type I offered clearadvantage for separate pathways of control for the two eyes. Thedistribution of the innervations signal to the two eyes is repre-sented by the diagram of Fig 8. The most commonly accepted

V condition and the saccadic amplitude in the BV condition for each subject and forn AEV vs. BV conditions and for each subject were calculated for two degree bins ofrvals were drawn for each eye (affected eye in blue and sound eye in red) in BV and

Fig. 8. The innervation results from an association between the pulse and the stepsignal. The distribution of the innervation signals to the two eyes is represented bythis diagram. A signal of pulse and step is directed towards each eye. The motorinput from the abducens motoneurons toward the affected eye is drawn as a dottedline. The step to the affected eye is adapted through gain K. The hypothesis for thelevel of adaptation through gain KAEV (drawn in grey) is illustrated in the diagram.There is possibility of adaptation for the pulse (gain = KAEV) signal coupled for thetwo eyes and the ability to adapt the step signal separately for each eye (gain K).Adapted from Yüksel et al. (2008).


hypothesis of conjugate horizontal eye movements guided by Her-ing’s law of equal innervations for extra ocular muscles seems to beconfirmed in DRS even though recent studies have found evidencethat some premotor neurons that should encode binocular com-mands may actually encode monocular commands (King & Zhou,2000; Zhou & King, 1998). Besides, some clinical observations de-scribe abnormalities producing disjunctive eye movements withinthe framework of independent control of the two eyes (Dell’Osso,1994). These results are in agreement with the observations re-ported in repair properties in peripheral and central 6th nerve pal-sies (Wong, McReelis, & Sharpe, 2006). Evidence has been giventhat monocular unidirectional change in saccade speed (pulse sig-nal of innervation) was possible in peripheral palsies but not incentral palsies. Explanation was proposed either by regenerationof axons/myelin peripherally or by mechanisms of monocularadaptation. As DRS is a congenital deficit in abducens motoneuronsat brainstem level, it seems probable that such mechanisms ofmonocular adaptation of the pulse signal of innervation are dam-aged. However, DRS offers compensation by separate adaptationof the step signal of innervation for the two eyes (Fig 8).

10. The interest for modeling saccades in DRS

The goal of theoretical studies on eye movements is to developmodels that realistically represent neurobiological processes.Although each aspect of a movement can be explained by manymodels, the requirement that a single model accounts for as much

normal and abnormal behavior as possible constrains the choice ofmodels and reveals isomorphisms that contribute to our under-standing of brain function. Thus, a key factor in modeling the neu-ral control of saccades is the interaction between clinical and basicscience.

Biomechanical modeling of saccades in DRS type I can have bothclinical utility and scientific validity. As a scientific hypothesis,such a model would propose an explanation for the pathogenesisof the oculomotor disorder. Indeed, all available histological, elec-tromyographic and imagery data of DRS brought explanation to thepathogenesis of DRS as being a hypoplasia of motoneurons of theabducens nucleus and nerve on the affected side with secondaryanomalous innervation of the affected lateral rectus muscle at aperipheral level by the branches of the oculomotor nerve. But, ifwe look from bottom-up, there is still uncertainty about the orga-nization of the oculomotor control at the brainstem level. Studyingeye movements is an elegant non-invasive way to collect experi-mental data on DRS. Binocular recordings give information onthe conjugacy of saccades. Moreover, monocular viewing conditionin binocular recordings gives insight to adaptation of the brain. Theparameterization of the properties of saccades is adapted to thepathological condition according to the knowledge of the proper-ties of the studied condition. This allows implementing a mathe-matical model. This model needs to take into account theabnormal innervation of the affected eye but also the changes inextra ocular muscles properties associated with the condition.The parameters of the model can be tuned on a subject-by-subjectbasis in order to gain insights into the affected eye plant and itsresidual innervation for each subject individually. All in all, themodel can result in a realistic and idiosyncratic representation ofthe neurophysiologic process involved in DRS. As a clinical tool,the model will help to assimilate the underlying pathogenic causeto the disorder with direct implication on therapeutical manage-ment. Indeed, additional quantitative information on the innerva-tional anomaly of the affected lateral rectus muscle in DRS wouldbe essential to guide surgical management. Electromyographyhas pitfalls and remains invasive for patient care. Imagery is notprecise enough to quantify the anomalous peripheral innervation.Studying saccades and modeling DRS is a promising way to guideclinicians in patient care management (Orban de Xivry, Yüksel, &Lefèvre, 2007).

11. New classification of DRS

Daily practice of medicine requires knowledge of all the clinicalsymptoms and signs characteristic of a pathology in order to matchinformation with the complaints of the patient. This leads to exactdiagnosis, sometimes after performing paraclinical tests to confirmclinical impression. Finally, diagnosis is followed by proposal ofappropriate treatment. This difficult task is made somewhat easierwith classification of pathological processes. Classification is usu-ally made based on clinical presentation.

General classification of DRS into types I–III was mainly basedon clinical features taking into account only the degree of asymme-try between the limitation of adduction and abduction (Von Noor-den, 2002). DRS type I was characterized by marked limitation orabsence of abduction; normal or only slightly defective adduction;narrowing of the palpebral fissure and retraction on adduction;widening of the palpebral fissure on attempted abduction. DRStype II was described with limitation or absence of adduction withexotropia of the affected eye; normal or slightly limited abduction;narrowing of the palpebral fissure and retraction of the globe onattempted adduction. DRS type III was a combination of limitationof both abduction and adduction; retraction of the globe and nar-rowing of the palpebral fissure on attempted adduction. Although


this terminology has been useful for many years, a particular atten-tion should be given to the complaints of the patient and to carefulexamination of the head posture, the facial asymmetry, the eyemotility and the palpebral fissures on a dynamical basis (throughdifferent positions of the head and the gaze). This leads to makea major difference between the subgroups of DRS type I regardingto the abnormal head posture and the misalignment of the eyes inthe primary position. Moreover, in view of all the available electro-myographic, imaging and eye movement recordings, a new classi-fication of DRS may be more relevant because it will allow linkingthe clinical description with the degree of innervation anomaly andemphasize the continuum that exists between the different formsof DRS (Souza-Dias, 2009).

Fig. 9A summarizes 6th nerve palsy and different subtypes ofDRS type I. The common clinical feature between complete abdu-

Fig. 9. New classification of various subtypes of DRS according to the underlying innervDRS type I. (B) Update of classification of DRS types II and III.

cens nerve palsy and DRS type I is mainly the limitation of abduc-tion which is present in various degrees regarding the amount ofhypoplasia of the involved abducens nerve in parallel with noneto some supply from the oculomotor nerve. The classification fromleft to right is organized in a degressive manner; marked limitationof abduction to almost normal abduction. The amount of fibers thatabandon the medial rectus nerve for supply to the lateral rectus isvariable; if they are scarce, the medial rectus is much stronger thanthe lateral rectus and, consequently, the adduction is not signifi-cantly impaired; the bridle effect produced by the co-contractionis minor, with consequent small retraction in adduction and noanomalous vertical deviations (up and downshoot). If the numberof fibers that abandon the medial rectus nerve toward the lateralrectus is greater, there is lesser difference in forces between themedial and the lateral recti in adduction, because the innervation

ational deficit. (A) Comparison between complete 6th nerve palsy and subgroups of


of the medial rectus is reduced and the one of the lateral rectus isincreased. The adduction starts to be somewhat impaired, the bri-dle effect is stronger, with larger retraction, and there can be smallanomalous vertical movements. This phenomenon leads to pro-gressive change in the compensatory abnormal head posture; turntoward the affected side, no torticollis, turn toward the sound side.Complete abducens nerve palsy (Fig. 9A, first column) representscomplete absence of residual innervation (=0) and no supply fromthe oculomotor nerve (=0). Eso-DRS type I (Fig. 9A, second column)is explained by some residual innervation of the abducens nerveand some supply from the oculomotor nerve in a ratio of 3rd nervesupply < 6th nerve residual innervation. Ortho-DRS type I (Fig. 9A,third column) represents a ratio of 6rd nerve supply = 6th nerveresidual innervation. Exo-DRS type I (Fig. 9A , fourth column) rep-resents a ratio of 3rd nerve supply > 6th nerve residual innervation.

Fig. 9B compares DRS types II, III and synergistic divergence.DRS types I and III are in fact a continuum as the number of fibersthat abandon the medial rectus nerve increases, which leads thismuscle to lose force and the lateral rectus to gain force in its abnor-mal contraction, until arriving to the situation in which their forcesequalize themselves (symmetric co-contraction). In this situationthere is no adduction or abduction (the abducens nerve is absent),the bridle effect of the co-contraction is maximal and consequentlythe retraction and the anomalous vertical movements are moreevident (Fig. 9B, first column). Therefore, it is more relevant to re-name type III as type II. The increase of the amount of residualinnervation of the abducens nerve leads to DRS type II that shouldbe renamed type III because the abduction becomes near to nor-mal (Fig. 9B, second column). The clinical motility deficit is in-versed with marked limitation of adduction greater than thelimitation of abduction. Moreover, this form of DRS must be partic-ularly distinguished because the presence of abducens nerve hasbeen shown to be preserved on MRI studies in some forms ofDRS (Kim & Hwang, 2005b). The final type is a variant of DRS typeII with simultaneous abduction, in fact an extreme form (Fig. 9B,third column). If the number of fibers that leave the medial rectusnerve toward the lateral rectus is still greater, one arrives to the sit-uation classically known as synergistic divergence. The clinicalpattern is fantastic. The lateral rectus becomes stronger than themedial rectus in the co-contraction and, consequently, in the at-tempt of adduction the affected eye abducts instead of adducting.When the sound eye abducts, the affected one also abducts obey-ing the Sherrington equal innervation law. It seems to be mainlya misdirection syndrome more than a basic maldevelopment ofthe oculomotor neurons. The basic etiology for this rare oculomo-tor disorder is still unknown.

The purpose of this new classification mainly aims at under-standing the underlying innervational abnormalities with correla-tion to the observed clinical pattern and to reflect the continuumbetween the different types of DRS. The quantification of theamount of residual innervation of the abducens nerve and theamount of supply from the oculomotor nerve can be obtained bymodeling saccades. In spite of all these considerations, these twotables give a guideline for approaching the different subtypes ofDRS which are often confused. Careful attention to the descriptionof the type of DRS is essential to interpret results of imaging or eyemovement recordings studies and is certainly very useful for guid-ing the clinician in the management of the patient.

Acknowledgments

This work was supported by the Fonds National de la RechercheScientifique, the Fondation pour la Recherche Scientifique Médi-cale, the Belgian Program on Interuniversity Attraction Poles initi-ated by the Belgian Federal Science Policy Office, Actions deRecherche Concertées (French community, Belgium), an internal

research grant (Fonds Spéciaux de Recherche) of the Universiteécatholique de Louvain, and the European Space Agency (ESA) ofthe European Union. JJO is supported by the Belgian American Edu-cational Foundation, by an internal research grant from the Univer-sité catholique de Louvain (Fonds spéciaux de recherche) and bythe Fondation pour la Vocation (Belgium).

References

Ahluwalia, B. K., Gupta, N. C., Goel, S. R., & Khurana, A. K. (1988). Study of Duane’sretraction syndrome. Acta Ophthalmologica (Copenhagen), 66(6), 728–730.

Akiko, Yamagami, Masato, Wakakura, Tetsuro, Tsubouchi, & Keichi, Goto (2003).Bilateral Duane’s retraction syndrome due to large pontine glioma. Neuro-Ophthalmology Japan, 20(1), 66–70.

Aoki, S., & Mukuno, K. (1989). Duane’s syndrome. Ganka (Ophthalmology), 31,523–530.

Appukuttan, B., Gillanders, E., Juo, S. H., Freas-Lutz, D., Ott, S., Sood, R., et al. (1999).Localization of a gene for Duane retraction syndrome to chromosome 2q31. TheAmerican Journal Human Genetics, 65(6), 1639–1646.

Barbe, M. E., Scott, W. E., & Kutschke, P. J. (2004). A simplified approach to thetreatment of Duane’s syndrome. British Journal of Ophthalmology, 88(1),131–138.

Blodi, F. C., Vanallen, M. W., & Yarbrough, J. C. (1964). Duane’s syndrome: A brainstem lesion. An electromyographic study. Archives of Ophthalmology, 72,171–177.

Breinin, G. M. (1957). Electromyography; a tool in ocular and neurologic diagnosis.II. Muscle palsies. AMA Archives of Ophthalmology, 57(2), 165–175.

Chung, M., Stout, J. T., & Borchert, M. S. (2000). Clinical diversity of hereditaryDuane’s retraction syndrome. Ophthalmology, 107(3), 500–503.

Clarke, A. H., Ditterich, J., Druen, K., Schonfeld, U., & Steineke, C. (2002). Using highframe rate CMOS sensors for three-dimensional eye tracking. Behavior ResearchMethods, Instruments, and Computers, 34(4), 549–560.

Collewijn, H., van der Mark, F., & Jansen, T. C. (1975). Precise recording of human eyemovements. Vision Research, 15(3), 447–450.

Dell’Osso, L. F. (1994). Evidence suggesting individual ocular motor control of eacheye (muscle). Journal of Vestibular Research, 4(5), 335–345.

Demer, J. L., Clark, R. A., Lim, K. H., & Engle, E. C. (2007a). Magnetic resonanceimaging evidence for widespread orbital dysinnervation in dominant Duane’sretraction syndrome linked to the DURS2 locus. Investigative Ophthalmology andVisual Science, 48(1), 194–202.

Demer, J. L., Clark, R. A., Lim, K. H., & Engle, E. C. (2007b). Magnetic resonanceimaging of innervational and extraocular muscle abnormalities in Duane-radialray syndrome. Investigative Ophthalmology and Visual Science, 48(12),5505–5511.

DeRespinis, P. A., Caputo, A. R., Wagner, R. S., & Guo, S. (1993). Duane’s retractionsyndrome. Survey of Ophthalmology, 38(3), 257–288.

Duane, A. (1905). Congenital deficiency of abduction, associated with impairment ofadduction, retraction movements, contraction of palpebral fissure and obliquemovements of the eye. Archives of Ophthalmology, 34, 133–159.

Duane, T. D., Schatz, N. J., & Caputo, A. R. (1976). Pseudo-Duane’s retractionsyndrome. Transactions of the American Ophthalmological Society, 74, 122–132.

Engle, E. C., Andrews, C., Law, K., & Demer, J. L. (2007). Two pedigrees segregatingDuane’s retraction syndrome as a dominant trait map to the DURS2 geneticlocus. Investigative Ophthalmology and Visual Science, 48(1), 189–193.

Evans, J. C., Frayling, T. M., Ellard, S., & Gutowski, N. J. (2000). Confirmation oflinkage of Duane’s syndrome and refinement of the disease locus to an 8.8-cMinterval on chromosome 2q31. Human Genetics, 106(6), 636–638.

Gobin, M. H. (1972). Aspects of horizontal muscle surgery, particularly in Duane’ssyndrome. Transactions of the Ophthalmological Societies, UK, 92, 685–693.

Gourdeau, A., Miller, N., Zee, D., & Morris, J. (1981). Central ocular motorabnormalities in Duane’s retraction syndrome. Archives of Ophthalmology,99(10), 1809–1810.

Hofmann, R. J. (1985). Monozygotic twins concordant for bilateral Duane’sretraction syndrome. American Journal of Ophthalmology, 99(5), 563–566.

Hotchkiss, M. G., Miller, N. R., Clark, A. W., & Green, W. R. (1980). Bilateral Duane’sretraction syndrome. A clinical-pathologic case report. Archives ofOphthalmology, 98(5), 870–874.

Hoyt, W. F., & Nachtigaller, H. (1965). Anomalies of ocular motor nerves.Neuroanatomic correlates of paradoxical innervation in Duane’s syndromeand related congenital ocular motor disorders. American Journal ofOphthalmology, 60(3), 443–448.

Huber, A. (1974). Electrophysiology of the retraction syndromes. British Journal ofOphthalmology, 58(3), 293–300.

Jampolsky. (1999). Duane syndrome. In: W.B.S.C. (Ed.), Clinical strabismusmanagement: Principles and surgical techniques (pp. 325–346.).

Kang, N. Y., & Demer, J. L. (2006). Comparison of orbital magnetic resonance imagingin Duane syndrome and abducens palsy. American Journal of Ophthalmology,142(5), 827–834.

Kapoula, Z., Bucci, M. P., Eggert, T., & Garraud, L. (1997). Impairment of the binocularcoordination of saccades in strabismus. Vision Research, 37(19), 2757–2766.

Kargi, S. H., Atasoy, H. T., Sayarlioglu, H., Koksal, M., Ugurbas, S. H., & Alpay, A.(2005). Acquired retraction of the eye as the first sign of myositis. Strabismus,13(2), 85–88.


Kaufman, L. W., Folk, E. R., & Miller, M. T. (1989). Monozygotic twins discordant forDuane’s retraction syndrome. Case report. Archives of Ophthalmology, 107(3),324–325.

Khan, A. O. (2005). Inverse globe retraction syndrome complicating recurrentpterygium. British Journal of Ophthalmology, 89(5), 640–641.

Khan, A. O., & Oystreck, D. T. (2006). Fixation preference for the affected eye inpatients with unilateral Duane syndrome. Journal AAPOS, 10(3), 275–276.

Khodadoust, A. A., & von Noorden, G. K. (1967). Bilateral vertical retractionsyndrome. A family study. Archives of Ophthalmology, 78(5), 606–612.

Kim, J. H., & Hwang, J. M. (2005a). Hypoplastic oculomotor nerve and absentabducens nerve in congenital fibrosis syndrome and synergistic divergencewith magnetic resonance imaging. Ophthalmology, 112(4), 728–732.

Kim, J. H., & Hwang, J. M. (2005b). Presence of the abducens nerve according to thetype of Duane’s retraction syndrome. Ophthalmology, 112(1), 109–113.

King, W. M., & Zhou, W. (2000). New ideas about binocular coordination of eyemovements: Is there a chameleon in the primate family tree? AnatomicalRecord, 261(4), 153–161.

Kirkham, T. H. (1969). Duane’s syndrome and familial perceptive deafness. BritishJournal of Ophthalmology, 53(5), 335–339.

Kirkham, T. H. (1970a). Duane’s retraction syndrome and cleft palate. AmericanJournal of Ophthalmology, 70(2), 209–212.

Kirkham, T. H. (1970b). Inheritance of Duane’s syndrome. British Journal ofOphthalmology, 54(5), 323–329.

Kivlin, J. D., & Lundergan, M. K. (1985). Acquired retraction syndrome associated withorbital metastasis. Journal of Pediatric Ophthalmology and Strabismus, 22(3),109–112.

Leigh, J. R., & Zee, D. S. (2006). The neurology of eye movements. (p. 776). New York:Oxford University Press Inc.

Magoon, E., Cruciger, M., Scott, A. B., & Jampolsky, A. (1982). Diagnostic injection ofXylocaine into extraocular muscles. Ophthalmology, 89(5), 489–491.

Matteuci, P. (1946). I difetti congenti di abduzione con particolare riguardo allapathogenesi. Rassegna Ital Ottalmologia, 15, 345–380.

Mehdorn, E., & Kommerell, G. (1979). Inherited Duane’s syndrome: Mirror-likelocalization of oculomotor disturbances in monozygotic twins. Journal ofPediatric Ophthalmology and Strabismus, 16(3), 152–155.

Metz, H. S. (1983). Saccadic velocity measurements in strabismus. Transactions ofthe American Ophthalmological Society, 81, 630–692.

Metz, H. S., Scott, A. B., & Scott, W. E. (1975). Horizontal saccadic velocities inDuane’s syndrome. American Journal of Ophthalmology, 80(5), 901–906.

Miller, J. M. (1989). Functional anatomy of normal human rectus muscles. VisionResearch, 29(2), 223–240.

Miller, N. R., Kiel, S. M., Green, W. R., & Clark, A. W. (1982). Unilateral Duane’sretraction syndrome (Type 1). Archives of Ophthalmology, 100(9),1468–1472.

Miyake, N., Chilton, J., Psatha, M., Cheng, L., Andrews, C., Chan, W. M., et al. (2008).Human CHN1 mutations hyperactivate alpha2-chimaerin and cause Duane’sretraction syndrome. Science, 321(5890), 839–843.

Mizukawa, K., Kimura, H., Fukai, S., & Tabuchi, A. (2004). Classification of Duane’sretraction syndrome: Two additional electromyogram types. Japanese Journal ofOphthalmology, 48(2), 148–153.

Moore, L. D., Feldon, S. E., & Liu, S. K. (1988). Infrared oculography of Duane’sretraction syndrome (type 1). Archives of Ophthalmology, 106(7), 943–946.

Moorman, C. M., & Elston, J. S. (1995). Acute orbital myositis. Eye, 9(Pt 1),96–101.

Murthy, R. (2008). Inverse Duane’s retraction syndrome following myocysticercosis.Indian Journal of Ophthalmology, 56(1), 89–90.

Nemet, P., & Ron, S. (1978). Ocular saccades in Duane’s syndrome. British Journal ofOphthalmology, 62(8), 528–532.

Orban de Xivry, J. -J., Yüksel, D., Lefèvre, P. (2007). A model of saccade adaptationbased on recordings from Duane retraction syndrome patients. Society forNeuroscience (abstract #167.5).

Osher, R. H., Schatz, N. J., & Duane, T. D. (1980). Acquired orbital retractionsyndrome. Archives of Ophthalmology, 98(10), 1798–1802.

Ozkan, S. B., & Aribal, M. E. (2007). Comparison of orbital magnetic resonanceimaging in duane syndrome and abducens palsy. American Journal ofOphthalmology, 143(5), 907 (author reply 907–908).

Ozkurt, H., Basak, M., Oral, Y., & Ozkurt, Y. (2003). Magnetic resonance imaging inDuane’s retraction syndrome. Journal of Pediatric Ophthalmology and Strabismus,40(1), 19–22.

Parsa, C. F., Grant, E., Dillon, W. P., Jr., du Lac, S., & Hoyt, W. F. (1998). Absence of theabducens nerve in Duane syndrome verified by magnetic resonance imaging.American Journal of Ophthalmology, 125(3), 399–401.

Pfaffenbach, D. D., Cross, H. E., & Kearns, T. P. (1972). Congenital anomalies inDuane’s retraction syndrome. Archives of Ophthalmology, 88(6), 635–639.

Piccirelli, M., Luechinger, R., Sturm, V., Boesiger, P., Landau, K., & Bergamin, O.(2009). Local deformation of extraocular muscles during eye movement.Investigative Ophthalmology and Visual Science, 50(11), 5189–5196.

Prakash, P., & Menon, V. (1981). Combined horizontal and vertical retractionsyndrome. Indian Journal of Ophthalmology, 29(2), 113–114.

Prieto-Diaz, J. (1985). Duane’s syndrome electro-oculography pattern. Journal ofPediatric Ophthalmology and Strabismus, 22(6), 243–245.

Robinson, D. A. (1963). A method of measuring eye movement using a scleral searchcoil in a magnetic field. IEEE Transactions on Bio-medical Engineering, BME-10,137–145.

Rosenbaum, A. L. (2004). Costenbader lecture. The efficacy of rectus muscletransposition surgery in esotropic Duane syndrome and VI nerve palsy. JournalAAPOS, 8(5), 409–419.

Rosenbaum, A. L., & Weiss, S. J. (1978). Monozygotic twins discordant for Duane’sretraction syndrome. Journal of Pediatric Ophthalmology and Strabismus, 15(6),359–361.

Scott, A. B. (1971). Active force tests in lateral rectus paralysis. Archives ofOphthalmology, 85, 397–404.

Scott, A. B. (1975). Force and velocity tests in strabismus. Transactions of theAmerican Academy Ophthalmology and Otolaryngology, 79, 727–732.

Scott, A. B., & Wong, G. Y. (1972). Duane’s syndrome. An electromyographic study.Archives of Ophthalmology, 87(2), 140–147.

Sevel, D., & Kassar, B. S. (1974). Bilateral Duane syndrome. Occurrence in threesuccessive generations. Archives of Ophthalmology, 91(6), 492–494.

Silverberg, M., & Demer, J. (2001). Duane’s syndrome with compressive denervationof the lateral rectus muscle. American Journal of Ophthalmology, 131(1),146–148.

Smith, J. L., & Damast, M. (1973). Acquired retraction syndrome after sixth nervepalsy. British Journal of Ophthalmology, 57(2), 110–114.

Sood, G. C., Srinath, B. S., & Krishnamurthy, G. (1975). Acquired Duane’s retractionsyndrome following Kronlein’s operation. Eye, Ear, Nose & Throat Monthly, 54(8),308–310.

Souza-Dias, C. R. D. (2009). New considerations about Duane’s syndrome. RevistaBrasileira de Oftalmologia, 68, 107–113.

Spielmann, A. (1988). Bilateral vertical retraction syndrome in horizontal bilateralDuane’s syndrome. Graefes Archive for Clinical and Experimental Ophthalmology,226(5), 425–427.

Stilling, J. (1887). Untersuchungen über die Entstehung der Kurzsichtigkeit. Wiesbaden,Bergmann, 13.

Tredici, T. D., & von Noorden, G. K. (1985). Are anisometropia and amblyopiacommon in Duane’s syndrome? Journal of Pediatric Ophthalmology andStrabismus, 22(1), 23–25.

Türk, S. (1896). Über Retraktionsbewegungen der Augen. Deutsche medicinischeWochenschrift, Berlin, 22: 199.

Von Noorden, G. K. (2002). Special forms of strabismus: Retraction syndrome(Duane syndrome). In: C. Mosby (Ed.), Binocular Vision and Ocular Motility (pp.458–466). St-Louis.

von Noorden, G. K., & Murray, E. (1986). Up- and downshoot in Duane’s retractionsyndrome. Journal of Pediatric Ophthalmology and Strabismus, 23(5), 212–215.

Wong, A. M., McReelis, K., & Sharpe, J. A. (2006). Saccade dynamics in peripheral vscentral sixth nerve palsies. Neurology, 66(9), 1390–1398.

Yüksel, D., Optican, L. M., & Lefevre, P. (2005). Properties of saccades in Duaneretraction syndrome. Investigative Ophthalmology and Visual Science, 46(9),3144–3151.

Yüksel, D., Orban de Xivry, J. J., & Lefevre, P. (2008). Binocular coordination ofsaccades in Duane retraction syndrome. Vision Research, 48(19), 1972–1979.

Zhou, W., & King, W. M. (1998). Premotor commands encode monocular eyemovements. Nature, 393(6686), 692–695.

review of the major findings about duane retraction ...duane retraction syndrome (drs) is a...

Documents